PX4-Autopilot/docs/en/msg_docs/AutotuneAttitudeControlStatus.md

pageClass

pageClass
is-wide-page

AutotuneAttitudeControlStatus (UORB message)

Autotune attitude control status.

This message is published by the fw_autotune_attitude_control and mc_autotune_attitude_control modules when the user engages autotune, and is subscribed to by the respective attitude controllers to command rate setpoints.

The rate_sp field is consumed by the controllers, while the remaining fields (model coefficients, gains, filters, and autotune state) are used for logging and debugging.

TOPICS: autotune_attitude_control_status

Fields

Name	Type	Unit [Frame]	Range/Enum	Description
timestamp	uint64	us		Time since system start
coeff	float32[5]			Coefficients of the identified discrete-time model
coeff_var	float32[5]			Coefficients' variance of the identified discrete-time model
fitness	float32			Fitness of the parameter estimate
innov	float32	rad/s		Innovation (residual error between model and measured output)
dt_model	float32	s		Model sample time used for identification
kc	float32			Proportional rate-loop gain (ideal form)
ki	float32			Integral rate-loop gain (ideal form)
kd	float32			Derivative rate-loop gain (ideal form)
kff	float32			Feedforward rate-loop gain
att_p	float32			Proportional attitude gain
rate_sp	float32[3]	rad/s		Rate setpoint commanded to the attitude controller.
u_filt	float32			Filtered input signal (normalized torque setpoint) used in system identification.
y_filt	float32	rad/s		Filtered output signal (angular velocity) used in system identification.
state	uint8		STATE	Current state of the autotune procedure.

Enums

STATE

Name	Type	Value	Description
STATE_IDLE	uint8	0	Idle (not running)
STATE_INIT	uint8	1	Initialize filters and setup
STATE_ROLL_AMPLITUDE_DETECTION	uint8	2	FW only: determine required excitation amplitude (roll)
STATE_ROLL	uint8	3	Roll-axis excitation and model identification
STATE_ROLL_PAUSE	uint8	4	Pause to return to level flight
STATE_PITCH_AMPLITUDE_DETECTION	uint8	5	FW only: determine required excitation amplitude (pitch)
STATE_PITCH	uint8	6	Pitch-axis excitation and model identification
STATE_PITCH_PAUSE	uint8	7	Pause to return to level flight
STATE_YAW_AMPLITUDE_DETECTION	uint8	8	FW only: determine required excitation amplitude (yaw)
STATE_YAW	uint8	9	Yaw-axis excitation and model identification
STATE_YAW_PAUSE	uint8	10	Pause to return to level flight
STATE_VERIFICATION	uint8	11	Verify model and candidate gains
STATE_APPLY	uint8	12	Apply gains
STATE_TEST	uint8	13	Test gains in closed-loop
STATE_COMPLETE	uint8	14	Tuning completed successfully
STATE_FAIL	uint8	15	Tuning failed (model invalid or controller unstable)
STATE_WAIT_FOR_DISARM	uint8	16	Waiting for disarm before finalizing

Source Message

Source file (GitHub)

::: details Click here to see original file

# Autotune attitude control status
#
# This message is published by the fw_autotune_attitude_control and mc_autotune_attitude_control modules when the user engages autotune,
# and is subscribed to by the respective attitude controllers to command rate setpoints.
#
# The rate_sp field is consumed by the controllers, while the remaining fields (model coefficients, gains, filters, and autotune state) are used for logging and debugging.

uint64 timestamp  # [us] Time since system start

float32[5] coeff      # [-] Coefficients of the identified discrete-time model
float32[5] coeff_var  # [-] Coefficients' variance of the identified discrete-time model
float32 fitness       # [-] Fitness of the parameter estimate
float32 innov         # [rad/s] Innovation (residual error between model and measured output)
float32 dt_model      # [s] Model sample time used for identification


float32 kc    # [-] Proportional rate-loop gain (ideal form)
float32 ki    # [-] Integral rate-loop gain (ideal form)
float32 kd    # [-] Derivative rate-loop gain (ideal form)
float32 kff   # [-] Feedforward rate-loop gain
float32 att_p # [-] Proportional attitude gain

float32[3] rate_sp # [rad/s] Rate setpoint commanded to the attitude controller.

float32 u_filt  # [-] Filtered input signal (normalized torque setpoint) used in system identification.
float32 y_filt  # [rad/s] Filtered output signal (angular velocity) used in system identification.

uint8 state  # [@enum STATE] Current state of the autotune procedure.
uint8 STATE_IDLE = 0                      # Idle (not running)
uint8 STATE_INIT = 1                      # Initialize filters and setup
uint8 STATE_ROLL_AMPLITUDE_DETECTION = 2  # FW only: determine required excitation amplitude (roll)
uint8 STATE_ROLL = 3                      # Roll-axis excitation and model identification
uint8 STATE_ROLL_PAUSE = 4                # Pause to return to level flight
uint8 STATE_PITCH_AMPLITUDE_DETECTION = 5 # FW only: determine required excitation amplitude (pitch)
uint8 STATE_PITCH = 6                     # Pitch-axis excitation and model identification
uint8 STATE_PITCH_PAUSE = 7               # Pause to return to level flight
uint8 STATE_YAW_AMPLITUDE_DETECTION = 8   # FW only: determine required excitation amplitude (yaw)
uint8 STATE_YAW = 9                       # Yaw-axis excitation and model identification
uint8 STATE_YAW_PAUSE = 10                # Pause to return to level flight
uint8 STATE_VERIFICATION = 11             # Verify model and candidate gains
uint8 STATE_APPLY = 12                    # Apply gains
uint8 STATE_TEST = 13                     # Test gains in closed-loop
uint8 STATE_COMPLETE = 14                 # Tuning completed successfully
uint8 STATE_FAIL = 15                     # Tuning failed (model invalid or controller unstable)
uint8 STATE_WAIT_FOR_DISARM = 16          # Waiting for disarm before finalizing

:::